This updates to mojo commit d259eb58aa59 and limits the roll script to
only pull in the parts of //mojo that are currently being used. More
stuff will be dropped in the future.
We still use SkPicture as our rasterization backend, but now we're uploading
our layer tree to C++. A future patch will push the layer tree deeper into the
system and we'll eventually switch backends.
Now we print Dart_GetError only when we don't have a stack trace. Also, improve
the usability of the sky_shell command line tool by defaulting to "packages"
for the package-root and not crashing when we don't have a package-root until
we actually try to load a package. Finally, remove console spam when we fail to
bind the observatory port.
Fixes#689
- Add Canvas.getSaveCount()
- Make RenderClipRect call context.paintChildWithClip instead of doing the clipping itself
- Make ClipLayer take a Rect instead of a Size
- Make PaintingContext.canvas read-only
- Add PaintingContext.paintChildWithClip()
- Minor rearrangings of code and style tweaks
Now we have the ability to draw multiple PictureLayers. We still squash all the
pictures together into a single SkPicture for simplicity. In the future, we'll
submit them to C++ separately and composite them on the GPU thread.
This patch converts sky.Paint to be a pure Dart object, which means we don't
need to open a weak handle to sky.Paint. Avoiding the weak handle reduces the
amount of weak handle callbacks we need to process at the end of GC and
therefore reduces jank.
This patch simplifies the SkPicture we generate for Skia. Instead of drawing
everything into a nested SkPicture, we now draw everything into the top-level
picture, which requires us to apply the device scale factor in Dart.
Turns out we weren't passing kComputeSaveLayerInfo_RecordFlag, which meant we
weren't computing the information necessary for the GrLayerHoister in Skia to
do its work.
This CL makes demo_launcher scrolling 3x faster on a Nexus 5.
This code was written from scratch for Sky by copying what
linux did, however that doesn't match with the expectations
of the surrounding code. Instead here I just copied what Blink
was doing around this time by looking at:
https://blink.lc/blink/tree/Source/platform/fonts/mac/FontPlatformDataMac.mm?id=f0ab9794a9d1240ce5a57e510fdb82dbc8337906
Blink has since diverged from this codepath, but this at least
makes our FontPlatformData match how the rest of the mac code
expects it to operate. This is all way more complicated than
we need.
@chinmaygarde
This is supposed to make Viktor's game faster, but it's not clear
to me that it actually does. I've left the code
using the dart version of drawAtlas for now until Viktor can
verify that it looks correct.
I also added a wrapper for SkFilterQuality in the process of
debugging SkCanvas.drawAtlas since all drawAtlas examples
in Skia use FilterQuality.low. The bug which blocked me for
so long turned out to be that SkCanvas.drawAtlas doesn't
draw anything if antialiasing is turned on.
Issue #138.
R=abarth@google.com
The first one is that we weren't setting up a
FontCachePurgePreventer during drawText. It's not clear
that this is the correct fix, since Blink doesn't have
this FontCachePurgePreventer here either, but it's also
possible that they would hit this same ASSERT and just
not care (since ASSERTs are disabled on clusterfuzz).
The second fix is making ExceptionState actually track
whether it has thrown an exception or not. The c++ code
was depending on this working in order to return early
from dom functions and not crash!
R=abarth@google.com
Remove all code relating to shadow trees, insertion points, shadow
boundaries, traversing composed trees, distribution, template
documents, custom elements, registering elements, element registries,
element factories, shadow roots, etc.
Remove the following features from the IDLs and from the binding
generators: CustomElementCallbacks, Reflect*, EventHandler.
Remove the CSS custom pseudo-element concept, since we no longer have
a UA style sheet worth talking about, no longer have shadow trees or
custom elements, no longer use pseudo-elements, and generally
therefore don't use this code at all.
